Mood Detection from Physical and Neurophysical Data Using Deep Learning Models

Table 7

Comparison with the state-of-the-art results.

Study	User count	Period	Collection method	Resource	Algorithm	Accuracy
[28]	44	Short term	Electrocardiogram sensors	Video	LDA	82.35
[23]	52	Short term	Electrocardiogram sensors	Video	Statistical Analysis	75.58
[29]	23	Short term	Electrocardiogram sensors	Video	Support vector machines	76.21
[59]	40	Short term	Electrocardiogram sensors	Video	Naïve Bayes	77.34
[60]	40	Short term	Electrocardiogram sensors	Video	K-Nearest Neighbors	74.80
[61]	58	Short term	Electrocardiogram sensors	Video	Naïve Bayes	75.39
[62]	25	Short term	Electrocardiogram sensors	Video	Support Vector Machines	73.96
[63]	30	Short term	Electrocardiogram sensors	Pictures/video	Support Vector Machines	72.47
[24]	83	30 days	Body temperature sensors	Sensor	ANN/DL	73.56
[64]	42	Short term	Electrocardiogram sensors	Sensor	Support Vector Machines	80.75
[65]	10	Short term	Self-reporting/sensors	Sensor	CNN	81.00
Our study	15	365 days	Motion, heartbeat sensors/keystroke patterns	Sensor	Decision Tree	82.03
Our study	15	365 days	Motion, heartbeat sensors/keystroke patterns	Sensor	CNN	84.31